Extracting feature sets, Classification using NaiveBayes, Pickle basics

TextClassify.py

##Text classification is always in the form of binary.
##ie, either spam or not spam

import nltk
import random
from nltk.corpus import movie_reviews
import pickle

## Create a list of tuples or features

documents = [(list(movie_reviews.words(fileid)), category)
             for category in movie_reviews.categories()
             for fileid in movie_reviews.fileids(category)]

##documents = []

##for category in movie_reviews.categories():
##    for fileid in movie_reviews.fileids(category):
##        documents.append(list(movie_reviews.words(fileid)), category)


random.shuffle(documents)

##print(documents[1])

##Convert everything to lower case
all_words = []
for w in movie_reviews.words():
    all_words.append(w.lower())

##nltk freq distribution
all_words = nltk.FreqDist(all_words)
##print(all_words.most_common(15))
##print(all_words["stupid"])

##top 3,000 most common words
word_features = list(all_words.keys())[:3000]

##Find the words and mark their pressence as positive or negative
def find_features(document):
    words = set(document)
    features = {}
    for w in word_features:
        features[w] = (w in words)

    return features

##Word which is true implies that it is common to appear in negative reviews
##print((find_features(movie_reviews.words('neg/cv000_29416.txt'))))

##for all of our documents, saving the feature existence booleans and their
##respective positive or negative categories
featuresets = [(find_features(rev), category) for (rev, category) in documents]

# set that we'll train our classifier with
training_set = featuresets[:1900]

# set that we'll test against.
testing_set = featuresets[1900:]

##classifier = nltk.NaiveBayesClassifier.train(training_set)

##Calling the pickle file and using the saved classifier
classifier_f = open("naivebayes.pickle", "rb")
classifier = pickle.load(classifier_f)
classifier_f.close()

##Saving the classifiers as pickle object
##save_classifier = open("naivebayes.pickle","wb")
##pickle.dump(classifier, save_classifier)
##save_classifier.close()

print("Classifier accuracy percent:",(nltk.classify.accuracy(classifier, testing_set))*100)

classifier.show_most_informative_features(15)